Via the Register, here’s a pretty detailed look (from July 2007) at a BT experiment with ISP-based ad-targeter Phorm. The fascinating bit is that the end user can see Phorm’s involvement at all – if I were designing an advertising system in partnership with the ISPs, I’d make sure my involvement wasn’t detectable, and I certainly wouldn’t inject a hidden IFRAME into the page!
If I had to guess, I’d guess that this sort of tomfoolery is – ironically – designed explicitly to keep Phorm from having to work directly with the user’s IP address and to provide a meaningful mechanism for user opt-outs. And I’m very curious to see how that goes, since most ad targeting companies provide an opt-out but do their damnedest to hide it. A Phorm representative states in this article that users will get a full-page notification when the system’s first deployed (allowing them to easily opt-out) as well as notifications alongside behaviorally-targeted banners (again allowing them to easily opt-out.) If that’s the case, then I predict a very large percentage of opt-outs indeed, because the standard line – that behaviorally-targeted advertisements benefit consumers by making ads more relevant – doesn’t hold true for display advertising. The pool of advertisers just isn’t deep enough, and the behavioral targeting just isn’t precise enough. If in the end, Phorm tanks, it’ll probably be because they were too transparent.
{ 0 comments… add one now }